Immunity
When the immune system is challenged by a foreign antigen it responds by launching a protective response. This response is characterized by the coordinated interaction of both the innate and acquired immune systems. These systems, once thought to be separate and independent, are now recognized as two interdependent parts that when integrated fulfill two mutually exclusive requirements: speed (contributed by the innate system) and specificity (contributed by the adaptive system).
The innate immune system serves as the first line of defence against invading pathogens, holding the pathogen in check while the adaptive responses are matured. It is triggered within minutes of infection in an antigen-independent fashion, responding to broadly conserved patterns in the pathogens (though it is not non-specific, and can distinguish between self and pathogens). Crucially, it also generates the inflammatory and co-stimulatory milieu (sometimes referred to as the danger signal) that potentiates the adaptive immune system and steers (or polarizes it) towards the cellular or humoral responses most appropriate for combating the infectious agent (discussed in more detail below).
The adaptive response becomes effective over days or weeks, but ultimately provides the fine antigenic specificity required for complete elimination of the pathogen and the generation of immunologic memory. It is mediated principally by T and B cells that have undergone germline gene rearrangement and are characterized by an exquisite specificity and long-lasting memory. However, it also involves the recruitment of elements of the innate immune system, including professional phagocytes (macrophages, neutrophils etc.) and granulocytes (basophils, eosinophils etc.) that engulf bacteria and even relatively large protozoal parasites. Once an adaptive immune response has matured, subsequent exposure to the pathogen results in its rapid elimination (usually before symptoms of infection become manifest) because highly specific memory cells have been generated that are rapidly activated upon subsequent exposure to their cognate antigen.
Interdependence of Innate and Adaptive Responses
It is now thought that the earliest events following pathogen invasion are effected by cellular components of the innate immune system. The response is initiated when resident tissue macrophages and dendritic cells (DCs) encounter pathogen and become activated by signals generated by interaction between pattern-recognition receptors (PRRs) and the pathogen-associated molecular patterns (PAMPs) shared by large groups of microorganisms. The activated macrophages and DCs are stimulated to release various cytokines (including the chemokines IL-8, MIP-1α and MIP-1β), which constitute the danger signal and triggers an influx of Natural Killer (NK) cells, macrophages, immature dendritic cells into the tissues.
Loaded with antigen, the activated DCs then migrate to lymph nodes. Once there, they activate immune cells of the adaptive response (principally naïve B- and T-cells) by acting as antigen-presenting cells (APCs). The activated cells then migrate to the sites of infection (guided by the “danger signal”) and once there further amplify the response by recruiting cells of the innate immune system (including eosinophils, basophils, monocytes, NK cells and granulocytes). This cellular trafficking is orchestrated by a large array of cytokines (particularly those of the chemokine subgroup) and involves immune cells of many different types and tissue sources (for a review, see Luster (2002), Current Opinion in Immunology 14: 129-135).
Polarization of the Adaptive Immune Response
The adaptive immune response is principally effected via two independent limbs: cell-mediated (type 1) immunity and antibody-mediated or humoral (type 2) immunity.
Type 1 immunity involves the activation of T-lymphocytes that either act upon infected cells bearing foreign antigens or stimulate other cells to act upon infected cells. This branch of the immune system therefore effectively contains and kills cells that are cancerous or infected with pathogens (particularly viruses). Type 2 immunity involves the generation of antibodies to foreign antigens by B-lymphocytes. This antibody-mediated branch of the immune system attacks and effectively neutralizes extracellular foreign antigens.
Both limbs of the immune system are important in fighting disease and there is an increasing realization that the type of immune response is just as important as its intensity or its duration. Moreover, since the type 1 and type 2 responses are not necessarily mutually exclusive (in many circumstances an effective immune response requires that both occur in parallel), the balance of the type 1/type 2 response (also referred to as the Th1:Th2 response ratio/balance by reference to the distinct cytokine and effector cell subsets involved in the regulation of each response—see below) may also play a role in determining the effectiveness (and repercussions) of the immune defence.
In many circumstances the immune response is skewed heavily towards a type 1 or type 2 response soon after exposure to antigen. The mechanism of this type 1/type 2 skewing or polarization is not yet fully understood, but is known to involve a complex system of cell-mediated chemical messengers (cytokines, and particularly chemokines) in which the type 1/type 2 polarization (or balance) is determined, at least in part, by the nature of the initial PRR-PAMP interaction when the DCs and macrophages of the innate immune system are first stimulated and subsequently by the cytokine milieu in which antigen priming of naïve helper T cells occurs.
Two cytokines in particular appear to have early roles in determining the path of the immune response. Interleukin-12 (IL-12), secreted by macrophages, drives the type 1 response by stimulating the differentiation of Th1 cells, the helper cells that oversee the type 1 response. Another macrophage cytokine, interleukin-10 (IL-10) inhibits this response, instead driving a type 2 response.
The type 1 and type 2 responses can be distinguished inter alia on the basis of certain phenotypic changes attendant on priming and subsequent polarization of naïve helper T cells. These phenotypic changes are characterized, at least in part, by the nature of the cytokines secreted by the polarized helper T cells.
Th1 cells produce so-called Th1 cytokines, which include one or more of TNF, IL-1, IL-7, IFN-gamma, IL-12 and/or IL-18. The Th1 cytokines are involved in macrophage activation and Th1 cells orchestrate Type 1 responses. In contrast, Th2 cells produce so-called Th2 cytokines, which include one or more of IL-4, IL-5, IL-10 and IL-13. The Th2 cytokines promote the production of various antibodies and can suppress the type 1 response.
The involvement of Th1 and Th2 cells and cytokines in type 1:type 2 immune response polarization has given rise to the terms Th1 response and Th2 response being used to define the type 1 and type 2 immune responses, respectively. Thus, these terms are used interchangeably herein.
There is an increasing realization that the type of immune response is just as important in therapy and prophylaxis as its intensity or its duration. For example, an excess Th1 response can result in autoimmune disease, inappropriate inflammatory responses and transplant rejection. An excess Th2 response can lead to allergies and asthma. Moreover, a perturbation in the Th1:Th2 ratio is symptomatic of many immunological diseases and disorders, and the development of methods for altering the Th1:Th2 ratio is now a priority.
Alkaloids
The term alkaloid is used herein sensu stricto to define any basic, organic, nitrogenous compound which occurs naturally in an organism. The term alkaloid is also used herein sensu lato to define a broader grouping of compounds which include not only the naturally occurring alkaloids, but also their synthetic and semi-synthetic analogues and derivatives.
Most known alkaloids are phytochemicals, present as secondary metabolites in plant tissues (where they may play a role in defence), but some occur as secondary metabolites in the tissues of animals, microorganisms and fungi. There is growing evidence that the standard techniques for screening microbial cultures are inappropriate for detecting many classes of alkaloids (particularly highly polar alkaloids, see below) and that microbes (including bacteria and fungi, particularly the filamentous representatives) will prove to be an important source of alkaloids as screening techniques become more sophisticated.
Structurally, alkaloids exhibit great diversity. Many alkaloids are small molecules, with molecular weights below 250 Daltons. The skeletons may be derived from amino acids, though some are derived from other groups (such as steroids). Others can be considered as sugar analogues. It is becoming apparent (see Watson et al. (2001) Phytochemistry 56: 265-295) that the water soluble fractions of medicinal plants and microbial cultures contain many interesting novel polar alkaloids, including many carbohydrate analogues. Such analogues include a rapidly growing number of polyhydroxylated alkaloids.
Most alkaloids are classified structurally on the basis of the configuration of the N-heterocycle. Examples of some important alkaloids and their structures are set out in Kutchan (1995) The Plant Cell 7:1059-1070.
Watson et al. (2001) Phytochemistry 56: 265-295 have classified a comprehensive range of polyhydroxylated alkaloids inter alia as piperidine, pyrroline, pyrrolidine, pyrrolizidine, indolizidine and nortropanes alkaloids (see FIGS. 1-7 of Watson et al. (2001), the disclosure of which is incorporated herein by reference).
Watson et al. (2001), ibidem also show that a functional classification of at least some alkaloids is possible on the basis of their glycosidase inhibitory profile: many polyhydroxylated alkaloids are potent and highly selective glycosidase inhibitors. These alkaloids can mimic the number, position and configuration of hydroxyl groups present in pyranosyl or furanosyl moieties and so bind to the active site of a cognate glycosidase, thereby inhibiting it. This area is reviewed in Legler (1990) Adv. Carbohydr. Chem. Biochem. 48: 319-384 and in Asano et al., (1995) J. Med. Chem. 38: 2349-2356.
It has long been recognized that many alkaloids are pharmacologically active, and humans have been using alkaloids (typically in the form of plant extracts) as poisons, narcotics, stimulants and medicines for thousands of years. The therapeutic applications of polyhydroxylated alkaloids have been comprehensively reviewed in Watson et al. (2001), ibidem: applications include cancer therapy, immune stimulation, the treatment of diabetes, the treatment of infections (especially viral infections), therapy of glycosphingolipid lysosomal storage diseases and the treatment of autoimmune disorders (such as arthritis and sclerosis).
Both natural and synthetic mono- and bi-cyclic nitrogen analogues of carbohydrates are known to have potential as chemotherapeutic agents. Alexine (1) and australine (2) were the first pyrrolizidine alkaloids to be isolated with a carbon substituent at C-3, rather than the more common C-1 substituents characteristic of the necine family of pyrrolizidines.

The alexines occur in all species of the genus Alexa and also in the related species Castanospermum australe. Stereoisomers of alexine, including 1,7a-diepialexine (3), have also been isolated (Nash et al. (1990) Phytochemistry (29) 111) and synthesised (Choi et al. (1991) Tetrahedron Letters (32) 5517 and Denmark and Cottell (2001) J. Org. Chem. (66) 4276-4284).

Because of the reported weak in vitro antiviral properties of one 7,7a-diepialexine (subsequently defined as 1,7a-diepialexine), there has been some interest in the isolation of the natural products and the synthesis of analogues.
As an indolizidine alkaloid (and so structurally distinct from the pyrrolizidine alexines), swainsonine (4) is a potent and specific inhibitor of α-mannosidase and is reported to have potential as an antimetastic, tumour anti-proliferative and immunoregulatory agent (see e.g. U.S. Pat. No. 5,650,413, WO00/37465, WO93/09117).

The effect of variation in the size of the six-membered ring of swainsonine on its glycosidase inhibitory activity has been studied: pyrrolizidine derivatives (so-called “ring contracted swainsonines”) have been synthesised. However, these synthetic derivatives (1S,2R,7R,7aR)1,2,7-trihydroxypyrrolizidine (5) and the 7S-epimer (6)) were shown to have much weaker inhibitory activity relative to swainsonine itself (see U.S. Pat. No. 5,075,457).

Another compound, 1α,2α,6α,7α,7αβ-1,2,6,7-tetrahydroxypyrrolizidine (7) is an analogue of 1,8-diepiswainsonine and described as a “useful” inhibitor of glycosidase enzymes in EP0417059.

Casuarine, (1R,2R,3R,6S,7S,7aR)-3-(hydroxymethyl)-1,2,6,7-tetrahydroxypyrrolizidine (8) is a highly oxygenated bicyclic pyrrolizidine alkaloid that can be regarded as a more highly oxygenated analogue of the 1,7a-diepialexine (shown in 3) or as a C(3) hydroxymethyl-substituted analogue of the 1α,2α,6α,7α,7αβ-1,2,6,7-tetrahydroxypyrrolizidine (shown in 7).

Casuarine can be isolated from several botanical sources, including the bark of Casuarina equisetifolia (Casuarinaceae), the leaves and bark of Eugenia janibolana (Myrtaceae) and Syzygium guineense (Myrtaceae) (see e.g. Nash et al. (1994) Tetrahedron Letters (35) 7849-7852). Epimers of casuarine, and probably casuarine itself, can be synthesised by sodium hydrogen telluride-induced cyclisation of azidodimesylates (Bell et al. (1997) Tetrahedron Letters (38) 5869-5872).
Casuarina equisetifolia wood, bark and leaves have been claimed to be useful against diarrhoea, dysentery and colic (Chopra et al. (1956) Glossary of Indian Medicinal Plants, Council of Scientific and Industrial Research (India), New Delhi, p. 55) and a sample of bark has recently been prescribed in Western Samoa for the treatment of breast cancer. An African plant containing casuarine (identified as Syzygium guineense) has been reported to be beneficial in the treatment of AIDS patients (see Wormald et al. (1996) Carbohydrate Letters (2) 169-174).
The casuarine-6-α-glucoside (casuarine-6-α-D-glucopyranose, 9) has also been isolated from the bark and leaves of Eugenia jambolana (Wormald et al. (1996) Carbohydrate Letters (2) 169-174).

Eugenia jambolana is a well known tree in India for the therapeutic value of its seeds, leaves and fruit against diabetes and bacterial infections. Its fruit have been shown to reduce blood sugar levels in humans and aqueous extracts of the bark are claimed to affect glycogenolysis and glycogen storage in animals (Wormald et al. (1996) Carbohydrate Letters (2) 169-174).
Dendritic Cells and their Immunotherapeutic Uses
(a) Introduction
Dendritic cells (DCs) are a heterogeneous cell population with distinctive morphology and a widespread tissue distribution (see Steinman (1991) Ann. Rev. Immunol. 9: 271-296). They play an important role in antigen presentation, capturing and processing antigens into peptides and then presenting them (together with components of the MHC) to T cells. T cell activation may then be mediated by the expression of important cell surface molecules, such as high levels of MHC class I and II molecules, adhesion molecules, and costimulatory molecules.
Dendritic cells therefore act as highly specialized antigen-presenting cells (APCs): serving as “nature's adjuvants”, they potentiate adaptive T-cell dependent immunity as well as triggering the natural killer (NK and NKT) cells of the innate immune system. Dendritic cells therefore play a fundamental and important regulatory role in the magnitude, quality, and memory of the immune response. As a result, there is now a growing interest in the use of dendritic cells in various immunomodulatory interventions, which are described in more detail below.
Dendritic cells can be classified into different subsets inter alia on the basis of their state of maturation (mature or immature) and their cellular developmental origin (ontogeny). Each of these subsets appear to play distinct roles in vivo, as described below.
(b) Dendritic Cell Maturation
Immature (or resting) DCs are located in non-lymphoid tissue, such as the skin and mucosae, are highly phagocytic and readily internalize soluble and particulate antigens. It is only when such antigen-loaded immature DCs are also subject to inflammatory stimuli (referred to as maturation stimuli) that they undergo a maturation process that transforms them from phagocytic and migratory cells into non-phagocytic, highly efficient stimulators of naïve T cells.
Immature DCs are characterized by high intracellular MHC II in the form of MIICs, the expression of CD1a, active endocytosis for certain particulates and proteins, presence of FcgR and active phagocytosis, deficient T cell sensitization in vitro, low/absent adhesive and costimulatory molecules (CD40/54/58/80/86), low/absent CD25, CD83, p55, DEC-205, 2A1antigen, responsiveness to GM-CSF, but not M-CSF and G-CSF and a sensitivity to IL-10, which inhibits maturation.
Upon maturation, mature DCs, loaded with antigen and capable of priming T cells, migrate from the non-lymphoid tissues to the lymph nodes or spleen, where they process the antigen load and present it to the resident naïve CD4+ T cells and CD8+ cytotoxic T cells. This latter interaction generates CTLs, the cellular arm of the adaptive immune response, and these cells eliminate virally infected cells and tumour cells. The naïve CD4+ T cells differentiate into memory helper T cells, which support the differentiation and expansion of CD8+ CTLs and B cells. Thus, helper T cells exert anti-tumour activity indirectly through the activation of important effector cells such as macrophages and CTLs.
Having activated the T cells in this way, the mature DCs undergo apoptosis within 9-10 days.
Mature DC cells are characterized morphologically by motility and the presence of numerous processes (veils or dendrites). They are competent for antigen capture and presentation (exhibiting high MHC class I and II expression) and express a wide range of molecules involved in T cell binding and costimulation, (e.g. CD40, CD54/ICAM-1, CD58/LFA-3, CD80/B7-1 and CD86/B7-2) as well as various cytokines (including IL-12). They are phenotypically stable: there is no reversion/conversion to macrophages or lymphocytes.
Thus, mature DCs play an important role in T cell activation and cell-mediated immunity. In contrast, immature DCs are involved in regulating and maintaining immunological tolerance (inducing antigen-specific T cell energy).
(c) Dendritic Cell Ontogenic Subsets
Dendritic cells are not represented by a single cell type, but rather comprise a heterogeneous collection of different classes of cells, each with a distinct ontogeny. At least three different developmental pathways have been described, each emerging from unique progenitors and driven by particular cytokine combinations to DC subsets with distinct and specialized functions.
At present it is thought that the earliest DC progenitors/precursors common to all DCs originate in the bone marrow. These primitive progenitors are CD34+, and they are released from the bone marrow to circulate through both the blood and lymphoid organs.
Once released from the bone marrow, the primitive CD34+ DC progenitors are subject to various stimulatory signals. These signals can direct the progenitors along one of at least three different pathways, each differing with respect to intermediate stages, cytokine requirements, surface marker expression and biological function.                Lymphoid DCs are a distinct subset of DCs that are closely linked to the lymphocyte lineage. This lineage is characterized by the lack of the surface antigens CD11b, CD13, CD14 and CD33. Lymphoid DCs share ancestry with T and natural killer (NK) cells, the progenitors for all being located in the thymus and in the T cell areas of secondary lymphoid tissues. The differentiation of lymphoid DCs is driven by Interleukins 2, 3 and 15 (IL-3, IL-2 and IL-1), but not by granulocyte macrophage colony-stimulating factor (GM-CSF). Functionally, lymphoid promote negative selection in the thymus (possibly by inducing fas-mediated apoptosis) and are costimulatory for CD4+ and CD8+ T cells. More recently, lymphoid-like DCs derived from human progenitors have also been shown to preferentially activate the Th2 response. Because of their capacity to induce apoptosis and their role in eliminating potentially self-reactive T cells, it has been suggested that lymphoid DCs primarily mediate regulatory rather than stimulatory immune effector functions.        Myeloid DCs are distinguished by a development stage in which there is expression of certain features associated with phagocytes. There appear to be at least two structurally and functionally distinct subsets. The first is defined antigenically as CD14−, CD34+, CD68− and CD1a+ and sometimes referred to as DCs of the Langerhans cell type. This subset appears to prime T cells to preferentially activate Th1 responses and IL-12 appears implicated in this process. The subset may also activate naïve B cells to secrete IgM and may therefore be predominantly associated with an inflammatory Th1 response. A second myeloid DC subset, sometimes referred to as interstitial DCs, is defined antigenically as CD14+, CD68+ and CD1a− and related to monocytes (as a result they are also referred to as monocyte-derived DCs or Mo-DCs).(d) Dendritic Cell Vaccines        
In one dendritic cell-based treatment paradigm (reviewed in Schuler et al. (2003) Current Opinion in Immunol 15: 138-147), DC cells are taken from a patient (for example by apheresis) and then pulsed (primed or spiked) with a particular antigen or antigens (for example, tumour antigen(s)). They are then re-administered as an autologous cellular vaccine to potentiate an appropriate immune response.
In this treatment paradigm, the responding T cells include helper cells, especially Th1 CD4+ cells (which produce IFN-γ) and killer cells (especially CD8+ cytolytic T lymphocytes). The DCs may also mediate responses by other classes of lymphocytes (B, NK, and NKT cells). They may also elicit T cell memory, a critical goal of vaccination.
At present, little is known about the identity of the DC subset(s) required for optimum effectiveness of DC vaccines, beyond the recognition that maturation is required and immature DCs are to be avoided (Dhodapkar and Steinman (2002) Blood 100: 174-177).
Hsu et al. (1996) Nat Med 2: 52-58 used rare DCs isolated ex vivo from blood. These DCs were highly heterogeneous with respect to their ontogenic subsets but matured spontaneously during the isolation procedure. However, the yields were very low.
The yield problem has been addressed by the development of techniques for expanding the DCs ex vivo, for example with Flt3 ligand (Fong et al. (2001) PNAS 98: 8809-8814), but this is of limited effectiveness.
However, most studies have used Mo-DCs. These cells are obtained by exposing monocytes to GM-CSF and IL-4 (or IL-13) to produce immature Mo-DCs, which are then matured by incubation in a maturation medium. Such media comprise one or more maturation stimulation factor(s), and typically comprise Toll-like receptor (TLR) ligands (e.g. microbial products such as lipopolysaccharide and/or monophosphoryl lipid), inflammatory cytokines (such as TNF-α), CD40L, monocyte conditioned medium (MCM) or MCM mimic (which contains IL-1β, TNF-α, IL-6 and PGE2).
Although little is known at present about the influence of maturation medium on DC vaccine performance, MCM or MCM mimic currently represent a standard: Mo-DCs matured using these media are homogenous, have a high viability, migrate well to chemotactic stimuli and induce CTLs both in vitro and in vivo.
Techniques have been developed for generating large numbers of Mo-DCs (300 to 500 million mature DCs per apheresis) from adherent monocytes within semi-closed, multilayered communicating culture vessels offering a surface area large enough to cultivate one leuk-apheresis product. These so-called cell factories can be used to produce cryopreserved aliquots of antigen preloaded DCs which are highly viable on thawing, and optimised maturation and freezing procedures have been described (Berger et al. (2002) J. Immunol. Methods 268: 131-140; Tuyaerts et al. (2002) J. Immunol. Methods 264: 135-151).
Dendritic cells for vaccination have also been prepared from CD34+-derived DCs comprising a mixture of interstitial and DCs of the Langerhans cell type. Some workers believe that the latter DC subset are more potent than Mo-DCs when used as DC vaccines.
With regard to antigen selection, various approaches have been used. Both defined and undefined antigens can be employed. The antigens can be xenoantigens or autoantigens. One or more defined neoantigen(s) may be selected: in the case of cancer treatment, the neoantigen(s) may comprise a tumour-associated antigen. However, most popular are 9-11 amino acid peptides containing defined antigens (either natural sequences or analogues designed for enhanced MHC binding): such antigens can be manufactured to good manufacturing practice (GMP) standard and are easily standardized.
Other approaches have employed antigens as immune complexes, which are delivered to Fc-receptor-bearing DCs and which results in the formation of both MHC class I and MHC class II peptide sequences. This offers the potential for inducing both CTLs and Th cells (Berlyn et al. (2001) Clin Immunol 101: 276-283).
Methods have also been developed for exploring the whole antigenic repertoire of any given tumour (or other target cell, such as a virally-infected cell). For example, DC-tumour cell hybrids have been successfully used to treat renal cell carcinoma (Kugler et al. (2000) 6: 332-336), but the hybrids are difficult to standardize and short-lived. Necrotic or apoptotic tumour cells have been used, as have various cellular lysates.
It appears that the selection of patient-specific antigens may be important in the treatment of at least some cancers, and antigens derived from fresh tumour cells rather than tumour cell lines or defined antigens may prove important (Dhodapkar et al. (2002) PNAS 99: 13009-13013).
As regards delivery of the selected antigen(s) to the DCs, various techniques are available. Since the number and quality of MHC-peptide complexes directly influences the immunogenicity of the DC, the antigen loading technique may prove critical to DC vaccine performance (van der Burg et al. (1996) J Immunol 156: 3308-3314). It seems that prolonged presentation of MHC-peptide complexes by the DCs enhances immunogenicity and so loading techniques which promote prolonged presentation may be important. This has been achieved by loading the DCs internally through the use of peptides linked to cell-penetrating moieties (Wang and Wang (2002) Nat Biotechnol 20: 149-154).
Antigens can also be loaded by transfecting the DCs with encoding nucleic acid (e.g. by electroporation) such that the antigens are expressed by the DC, processed and presented at the cell surface. This approach avoids the need for expensive GMP proteins and antibodies. RNA is preferred for this purpose, since it produces only transient expression (albeit sufficient for antigen processing) and avoids the potential problems associated with the integration of DNA and attendant long-term expression/mutagenesis. Such transfection techniques also permit exploration of the whole antigenic repertoire of a target cell by use of total or PCR-amplified tumour RNA.
There is some evidence that helper proteins (for example, keyhole limpet hemocyanin (KLH) and tetanus toxoid (TT)) can provide unspecific help for CTL induction (Lanzavecchia (1998) Nature 393: 413-414) and it may prove advantageous to pulse DC with such helper proteins prior to vaccination.
With regard to posology, the dose, frequency and route of DC vaccine administration have not yet been optimised in clinical trials. Clearly, the absolute number of cells administered will depend on the route of administration and effectiveness of migration after infusion. In this respect there are indications that intradermal or subcutaneous administration may be preferred for the development of Th1 responses, although direct intranodal delivery has been employed to circumvent the need for migration from the skin to the nodes (Nestle et al. (1998) Nat Med 4: 328-332).
Quite distinct from the antigen-pulsed DC vaccine paradigm described above is an approach in which dendritic cells secreting various chemokines are injected directly into tumours where they have been shown to prime T cells extranodally (Kirk et al. (2001) Cancer Res 61: 8794-8802). Thus, in another treatment paradigm, DCs are targeted to a tumour and activated to elicit immune responses in situ without the need for ex vivo antigen loading.
In situ DC vaccination constitutes yet another distinct (but related) approach (Hawiger et al. (2001) J Exp Med 194: 769-779. In this therapeutic paradigm, antigen is targeted to DCs in vivo which are expanded and induced to mature in situ. This approach depends on efficient targeting of antigen to endogenous DCs (for example, using exosomes—see Thery et al. (2002) Nat Rev Immunol 2: 569-579) and the development of maturation stimulants that can effectively trigger maturation (preferably of defined DC subset(s)) in vivo.
(e) Use of Dendritic Cells in Adoptive CTL Immunotherapy
Cytotoxic T lymphocytes (CTLs) can be administered to a patient in order to confer or supplement an immune response to a particular disease or infection (typically cancer). For example, tumour specific T cells can be extracted from a patient (e.g. by leukapheresis), selectively expanded (for example by tetramer-guided cloning—see Dunbar et al. (1999) J Immunol 162: 6959-6962) and then re-administered as an autologous cellular vaccine.
The clinical effectiveness, applicability and tractability of this type of passive immunotherapy can be greatly increased by using dendritic cells to prime the T cells in vitro prior to administration.
(f) Dendritic Cell-Based Approaches to the Treatment of Autoimmune Disorders
Dendritic cells are also involved in regulating and maintaining immunological tolerance: in the absence of maturation, the cells induce antigen-specific silencing or tolerance.
Thus, in another dendritic cell-based treatment paradigm, immature DCs are administered as part of an immunomodulatory intervention designed to combat autoimmune disorders. In such applications, the suppressive potential of the DCs has been enhanced by in vitro transfection with genes encoding cytokines.
(g) The Role of IL-2 in Dendritic Cell Function
Granucci et al. (2002) Trends in Immunol. 23: 169-171 have reported transient upregulation of mRNA transcripts for IL-2 in dendritic cells following microbial stimulus. In WO03012078 Granucci describes the important role played by DC-derived IL-2 in mediating not only T cell activation but also that of NK cells and goes on to suggest that DC-derived IL-2 is a key factor regulating and linking innate and adaptive immunity.
Moreover, systemic administration of IL-2 has recently been shown to enhance the therapeutic efficacy of a DC vaccine (Shimizu et al. (1999) PNAS 96: 2268-2273), while the presence of IL-2 was shown to be essential for specific peptide-mediated immunity mediated by dendritic cells in at least some DC vaccination regimes (Eggert et al. (2002) Eur J Immunol 32: 122-127). In their recent review, Schuler et al. (ibidem) conclude that “ . . . it might be worthwhile to explore the combination of DC vaccination with IL-2 administration, as the T-cell responses induced by DC vaccination appear enhanced and therapeutically more effective.”.
It will be clear from the foregoing discussion that dendritic cells are now proven as valuable tools in immunotherapy (particularly in the treatment of cancer), but that DC vaccination is still at a relatively early stage. Methods for preparing DCs are improving continuously and an increasing number of Phase I, II and III clinical trials are driving intense research and development in this area. However, there is still a need to improve efficacy at the level of DC biology.
The present inventors have now surprisingly discovered that casuarine and certain casuarine analogues have unexpected immunomodulatory activity, and that this activity may not be dependent on glycosidase inhibition.